Cell Tracking according to Biological Needs: Strong Mitosis-aware Multi-Hypothesis Tracker with Aleatoric Uncertainty

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)4826-4840
Seitenumfang15
FachzeitschriftIEEE Transactions on Medical Imaging
Jahrgang44
Ausgabenummer12
Frühes Online-Datum25 Juni 2025
PublikationsstatusVeröffentlicht - 2 Dez. 2025

Abstract

Cell tracking and segmentation enable biologists to extract insights from large-scale microscopy time-lapse data. Driven by local accuracy metrics, current tracking approaches often suffer from a lack of long-term consistency and an inability to correctly reconstruct lineage trees. To address this issue, we introduce a novel assignment strategy consisting of two key components. First, we propose an uncertainty estimation technique for motion estimation frameworks. This method relaxes single-point motion representations into probabilistic spatial densities using problem-specific test-time augmentations. Second, we leverage these spatial densities to define a novel mitosis-aware assignment problem formulation. This formulation allows multi-hypothesis trackers to model cell divisions and resolve false associations and mitosis detections based on long-term conflicts. Our framework integrates explicit biological knowledge into assignment costs and combines it with learned representations derived from spatial densities. We evaluate our approach on nine competitive datasets and demonstrate that it substantially outperforms the current state-of-the-art on biologically inspired metrics, achieving improvements by a factor of approximately six and providing new insights into the behavior of motion estimation uncertainty.

ASJC Scopus Sachgebiete

Zitieren

Cell Tracking according to Biological Needs: Strong Mitosis-aware Multi-Hypothesis Tracker with Aleatoric Uncertainty. / Kaiser, Timo; Schier, Maximilian; Rosenhahn, Bodo.
in: IEEE Transactions on Medical Imaging, Jahrgang 44, Nr. 12, 02.12.2025, S. 4826-4840.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Kaiser T, Schier M, Rosenhahn B. Cell Tracking according to Biological Needs: Strong Mitosis-aware Multi-Hypothesis Tracker with Aleatoric Uncertainty. IEEE Transactions on Medical Imaging. 2025 Dez 2;44(12):4826-4840. Epub 2025 Jun 25. doi: 10.1109/TMI.2025.3583148, 10.48550/arXiv.2403.15011
Download
@article{f6ad3e543b0a4d4d84a7e757167e038e,
title = "Cell Tracking according to Biological Needs: Strong Mitosis-aware Multi-Hypothesis Tracker with Aleatoric Uncertainty",
abstract = "Cell tracking and segmentation enable biologists to extract insights from large-scale microscopy time-lapse data. Driven by local accuracy metrics, current tracking approaches often suffer from a lack of long-term consistency and an inability to correctly reconstruct lineage trees. To address this issue, we introduce a novel assignment strategy consisting of two key components. First, we propose an uncertainty estimation technique for motion estimation frameworks. This method relaxes single-point motion representations into probabilistic spatial densities using problem-specific test-time augmentations. Second, we leverage these spatial densities to define a novel mitosis-aware assignment problem formulation. This formulation allows multi-hypothesis trackers to model cell divisions and resolve false associations and mitosis detections based on long-term conflicts. Our framework integrates explicit biological knowledge into assignment costs and combines it with learned representations derived from spatial densities. We evaluate our approach on nine competitive datasets and demonstrate that it substantially outperforms the current state-of-the-art on biologically inspired metrics, achieving improvements by a factor of approximately six and providing new insights into the behavior of motion estimation uncertainty.",
keywords = "Aleatoric Uncertainty, Cell Segmentation, Cell Tracking, Multi-Hypothesis Tracking",
author = "Timo Kaiser and Maximilian Schier and Bodo Rosenhahn",
note = "Publisher Copyright: {\textcopyright} 1982-2012 IEEE.",
year = "2025",
month = dec,
day = "2",
doi = "10.1109/TMI.2025.3583148",
language = "English",
volume = "44",
pages = "4826--4840",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

Download

TY - JOUR

T1 - Cell Tracking according to Biological Needs

T2 - Strong Mitosis-aware Multi-Hypothesis Tracker with Aleatoric Uncertainty

AU - Kaiser, Timo

AU - Schier, Maximilian

AU - Rosenhahn, Bodo

N1 - Publisher Copyright: © 1982-2012 IEEE.

PY - 2025/12/2

Y1 - 2025/12/2

N2 - Cell tracking and segmentation enable biologists to extract insights from large-scale microscopy time-lapse data. Driven by local accuracy metrics, current tracking approaches often suffer from a lack of long-term consistency and an inability to correctly reconstruct lineage trees. To address this issue, we introduce a novel assignment strategy consisting of two key components. First, we propose an uncertainty estimation technique for motion estimation frameworks. This method relaxes single-point motion representations into probabilistic spatial densities using problem-specific test-time augmentations. Second, we leverage these spatial densities to define a novel mitosis-aware assignment problem formulation. This formulation allows multi-hypothesis trackers to model cell divisions and resolve false associations and mitosis detections based on long-term conflicts. Our framework integrates explicit biological knowledge into assignment costs and combines it with learned representations derived from spatial densities. We evaluate our approach on nine competitive datasets and demonstrate that it substantially outperforms the current state-of-the-art on biologically inspired metrics, achieving improvements by a factor of approximately six and providing new insights into the behavior of motion estimation uncertainty.

AB - Cell tracking and segmentation enable biologists to extract insights from large-scale microscopy time-lapse data. Driven by local accuracy metrics, current tracking approaches often suffer from a lack of long-term consistency and an inability to correctly reconstruct lineage trees. To address this issue, we introduce a novel assignment strategy consisting of two key components. First, we propose an uncertainty estimation technique for motion estimation frameworks. This method relaxes single-point motion representations into probabilistic spatial densities using problem-specific test-time augmentations. Second, we leverage these spatial densities to define a novel mitosis-aware assignment problem formulation. This formulation allows multi-hypothesis trackers to model cell divisions and resolve false associations and mitosis detections based on long-term conflicts. Our framework integrates explicit biological knowledge into assignment costs and combines it with learned representations derived from spatial densities. We evaluate our approach on nine competitive datasets and demonstrate that it substantially outperforms the current state-of-the-art on biologically inspired metrics, achieving improvements by a factor of approximately six and providing new insights into the behavior of motion estimation uncertainty.

KW - Aleatoric Uncertainty

KW - Cell Segmentation

KW - Cell Tracking

KW - Multi-Hypothesis Tracking

UR - http://www.scopus.com/inward/record.url?scp=105009656355&partnerID=8YFLogxK

U2 - 10.1109/TMI.2025.3583148

DO - 10.1109/TMI.2025.3583148

M3 - Article

AN - SCOPUS:105009656355

VL - 44

SP - 4826

EP - 4840

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 12

ER -

Von denselben Autoren